There is a bit of a buzz at the moment that Google has come up with a ‘solution’ for those multiple ugly urls that come with dynamic websites.
The announcement back at the start of 2009 told us if we have a number of dynamically produced URLs for the same (or similar) content within our website, we can tell the search engines which URL we want them to ‘show’. Then more recently Google announced further that the canonical element can be used to Handle cross-domain content duplication.
When dealing with duplicate content within a website, an ugly dynamic URL might be http://domain.com/index.php?sid=abc123678345
Within the head of that page you would include this element:
<link rel=”canonical” href=”http://domain.com/blue-shoes.php”/>
This tells search engines (ie Google, Yahoo) that the preferred location of that content is actually http://domain.com/blue-shoes.php
Further info is available at Google Webmaster Central
The problem is…there are a number of recommendations an SEO will make for a site that is to be made ‘Search Engine Friendly’ and when you are working with developers with little understanding of SEO getting each of your recommendations implemented can be tough. One of the toughest topics is getting across the importance of dealing with duplicate content and I thought we were getting somewhere but the canonical element has just put us back a step, thanks Google!
You see, the canonical element is meant for situations where there is no other solution for dealing with duplicate content. But from the looks of things (and from recent experience) people are assuming it is an equal/easy alternative for 301 redirects. It is a band aid; Not a solution.
If you really read the posts at Google, they do say that it is a hint and that 301 redirects are still preferred:
Is rel=”canonical” a hint or a directive?
It’s a hint that we honor strongly. We’ll take your preference into account, in conjunction with other signals, when calculating the most relevant page to display in search results.
Where possible, the most important step is often to use appropriate 301 redirects.
Where 301 redirects are not possible, the rel=”canonical” link element can give us a better understanding of your site and of your preferred URLs.
Is rel=”canonical” a suggestion or a directive?
This new option lets site owners suggest the version of a page that Google should treat as canonical. Google will take this into account, in conjunction with other signals, when determining which URL sets contain identical content, and calculating the most relevant of these pages to display in search results.
Some more points:
- Even better is to not have links to multiple URLs for the same content at all, just the one.
- If you are submitting a Google xml sitemap, it shouldn’t contain URLs that are redirected.
- Just because you see other sites using it (apparently) successfuly; it doesn’t mean it will do for yours.
- No other solution? Really? Then make sure you know what you are doing!
Actually, it’s a directive, not a suggestion. Google, in particular, treats canonical URLs just like 301s (at a recent SMX panel, a Google engineer basically said it was the equivalent, not “sort of like it”). I also don’t see it as a bandaid at all, in the sense that it allows you to duplicate content in different places, including now on different domains (at least for Google). This is great for sites, like news and media sites, that have multiple properties they want to intentionally expose their content on. All while maintaining control over the true source/destination of the content.
Great thoughts though, and I agree that some (if not many) people will use it as a bandaid, instead of simply fixing their crappy information architecture and/or using 301 redirects.
Hye Jon, thanks for the comment. I do agree that this has a great many benefits but the main reason for the post was because I’m starting to see developers thinking it is an easy way out when producing a dynamic website when it isn’t.
I recently consulted on a new website (using OSC) that produced 4 different urls for product and category pages depending on where it was linked from. The developers solution was implementing the canonical tag (cos Google said so). I looked into the software myself and found out it was a simple check box within the admin which reduced the URLs to the one with the friendly structure. No need for canonical and hardly difficult to solve ‘the proper way’.
Another issue from the SEO perspective is that the canonical tag is for search engines, not people. We can still access those other urls (and link to them).
Still not convinced it’s equivalent to a 301 from what I’ve read about it. Would love to see where google has put that in writing
That definitely sucks, and I do think you’re making a really good point with your entry. And yeah, you’re correct in holding your breath on the “same as” statement until we see it in writing
Pingback: Weekly Search & Social News: 12/22/09 | Search Engine Journal
Pingback: Weekly Search & Social News: 12/22/09 « Site Seo Doctor Blog
Pingback: Weekly Search & Social News: 12/22/09 - SeoAlchemist - How to make gold with SEO.
Pingback: Weekly Search & Social News: 12/22/09 | Best Hosting Talk | Cheap Web Hosting Deals | Web Hosting Deals and Coupons